This is an old revision of the document!
The metafor package implements various meta-analytic models, methods, and techniques that have been described in the literature. The links below demonstrate how the models, methods, and techniques described in the respective references can be applied via the metafor package. The items are organized by topic (therefore, articles covering multiple topics may be listed multiple times). Alternatively, you can jump to the references at the bottom of the page for the list of articles in alphabetical order.
The following items correspond to books on meta-analysis. The analyses described in the books are reproduced using the metafor package.
The articles below cover the standard fixed-, random-, and mixed-effects (meta-regression) models for meta-analysis.
Multivariate/multilevel meta-analytic models can be used to account for non-independent sampling errors and/or true effects (e.g., due to the inclusion of multiple treatment studies, multiple endpoints, or other forms of clustering).
The conditional logistic model (also called hypergeometric-normal model) can be used to meta-analyze odds ratios (obtained from 2×2 table data).
The article below describes and illustrates Peto's (one-step) method for meta-analyzing (log) odds ratio.
The articles below describe the meta-analysis of proportions via various methods.
The articles below describe the meta-analysis of incidence rates and incidence rate ratios.
The use of the Mantel-Haenszel method for meta-analyzing risk differences, risk ratios, and odds ratios (for 2×2 table data) and for meta-analyzing incidence rate differences and incidence rate ratios (for two-group person-time data) is illustrated in the following article.
The article below discusses the calculation of effect size measures for pretest posttest control group designs.
Below are articles that compare various estimators for the amount of (residual) heterogeneity and/or describe methods for obtaining confidence intervals thereof.
The articles below illustrate/discuss the calculation of best linear unbiased predictions (BLUPs) (also called empirical Bayes estimates).
Instead of assuming normally distributed true effects, one can use mixture models to model heterogeneity in the true effects in a more flexible manner.
Berkey, C. S., Hoaglin, D. C., Antczak-Bouckoms, A., Mosteller, F., & Colditz, G. A. (1998). Meta-analysis of multiple outcomes by regression with random effects. Statistics in Medicine, 17(22), 2537-2550.
Gleser, L. J., & Olkin, I. (2009). Stochastically dependent effect sizes. In H. Cooper, L. V. Hedges, & J. C. Valentine (Eds.), The handbook of research synthesis and meta-analysis (2nd ed., pp. 357–376). New York: Russell Sage Foundation.
Raudenbush, S. W. (2009). Analyzing effect sizes: Random effects models. In H. Cooper, L. V. Hedges, & J. C. Valentine (Eds.), The handbook of research synthesis and meta-analysis (2nd ed., pp. 295–315). New York: Russell Sage Foundation.
Stijnen, T., Hamza, T. H., & Ozdemir, P. (2010). Random effects meta-analysis of event outcome in the framework of the generalized linear mixed model with applications in sparse data. Statistics in Medicine, 29(29), 3046-3067.
Yusuf, S., Peto, R., Lewis, J., Collins, R., & Sleight, P. (1985). Beta blockade during and after myocardial infarction: An overview of the randomized trials. Progress in Cardiovascular Disease, 27(5), 335-371.